Abstract The Postnatal Growth and Retinopathy of Prematurity (G-ROP) Study Group conducted two large-scale multi- center studies, sponsored by the National Eye Institute, to develop a prediction model on a retrospective cohort (n=7,483) and validate the prediction model on a prospective cohort (n=4,393) using postnatal weight gain to identify infants with severe ROP. The proposed project will perform secondary analyses of the comprehensive datasets collected by the G-ROP Group. The specific aims of this proposal do not overlap with the aims of the G-ROP studies, as the G-ROP Group has successfully developed a prediction model structured as modified ROP screening criteria that yield 100% sensitivity for treatment-requiring ROP and reduce infants receiving exams by 30%, and the model is being validated using data from the just- completed prospective cohort study. The proposed secondary analyses of the largest-available detailed ROP and related risk-factors datasets aim to achieve the following goals: (1) To develop and validate an exposure index for supplemental oxygen use, timing, and duration that is predictive of the incidence and severity of ROP. (2) To determine whether congenital heart disease is associated with the incidence and severity of ROP. (3) To evaluate whether maternal breast milk and donor breast milk feedings (timing and duration) are protective factors with regards to the incidence and severity of ROP. (4) To develop and validate an ROP risk score incorporating both traditional and novel risk factors for ROP risk stratification. (5) To develop and validate modified, risk-tiered dynamic eye examination schedules that minimize the number of eye examinations among infants who meet the G-ROP modified ROP screening criteria. The two similarly-designed G-ROP studies provide two unique datasets that can be used for model development and validation respectively, which improve the validity of the findings from these aims. The proposed secondary analyses may lead to better understanding of the risk factors for ROP and improved ROP risk stratification for further reducing the number of ROP exams by ophthalmologists. The very large sample size (N=11,876), high incidence rate of ROP (~42%) and of severe ROP (~12%), high quality of the datasets from diverse infants, as well as the rich experience and high productivity of the investigators of the G-ROP Study Group, support the feasibility and high likelihood of success of the proposed project.